The integumentary system, with skin as its largest organ, acts as a protective barrier against external agents while maintaining systemic homeostasis. Despite its resilience, the skin is susceptible to various conditions, such as injuries, infections, and chronic disorders, requiring accurate diagnostics for effective treatment. This paper presents a novel multi-modal deep learning framework for dermatological disease prediction and management. The proposed architecture integrates visual and textual data, employing the Inception V3 CNN for extracting features from dermal images and BERT for analysing clinical narratives. The combined embeddings are processed through a fully connected network for precise disease classification. The model, trained on a diverse annotated dataset, ensures applicability across varied skin types and conditions. Empirical results reveal that this hybrid approach outperforms traditional single-modality models in metrics like accuracy, precision, and recall. By providing detailed diagnostic insights, the framework aids in faster clinical decision-making, improved prognostics, and evidence-based treatments, enhancing dermatological care outcomes.